Pritchett, R., Minnis, H., Puckering, C., Rajendran, G., and Wilson, P. (2013) Can behaviour during immunisation be used to identify attachment patterns? A feasibility study. International Journal of Nursing Studies, 50 (3). pp. 386-391. ISSN 0020-7489
Copyright © 2013 Elsevier.
A copy can be downloaded for personal non-commercial research or study, without prior permission or charge The content must not be changed in any way or reproduced in any format or medium without the formal permission of the copyright holder(s) When referring to this work, full bibliographic details must be given http://eprints.gla.ac.uk/70161/ Deposited on: 3 March 2014
Enlighten – Research publications by members of the University of Glasgow http://eprints.gla.ac.uk
Can behaviour during immunisation be used to identify attachment patterns?
A feasibility study.
Abstract
Background Infant attachment is a strong predictor of mental health, and current
measures involve placing children into a stressful situation in order to observe how the
child uses their primary caregiver to assuage their distress.
Objectives This study aimed to explore observational correlates of attachment patterns
during immunisation.
Participants and Setting 18 parent child pairs were included in the study. They were
all recruited through a single GP practice.
Methods Infant immunisation videos were observed and coded for parenting behaviours
as well as pain promoting and pain reducing strategies. These scores were compared
between different attachment groups, as measured with the Manchester Child
Attachment Story Task.
Results Parents of securely attached children scored higher on positive Mellow
Parenting Observational System behaviours, but not at a statistically significant level.
Parents of securely attached children were also significantly more likely to engage in
pain reducing behaviours (p<0.01) than parents of insecurely attached children.
2
Conclusions Future research should go on to develop robust, composite measures for
attachment informative behaviours in the immunisation situation and test these in a fully
powered study.
Keywords; attachment, immunisation, infant, observation, primary care.
What is already known about the topic?
There is a need to develop a new attachment measure because there is no current tool
suitable for use in non-specialist settings (e.g. in primary care or paediatric clinics)
particularly, as in primary care consultations, where there are significant time
constraints.
What this paper adds?
This paper is a proof of the concept that immunisation has the potential to be used as an
attachment measure which has great clinical potential. Attachment could be classified
without placing a child into an artificial state of distress, as well as providing a
measurement which could be used with children of different ages. This gives the
potential for earlier diagnosis and treatment of attachment problems.
3
Attachment
Sroufe (Sroufe 2005) argued that “Nothing can be assessed in infancy that is more
important” [than attachment]. Attachment forms the basis of infant responses to
separation from their caregiver (Bowlby 1960), how they respond to strangers (Morgan
and Ricciuti 1969;Schaffer 1966) and how freely they explore their environment
(Ainsworth 1967). Insecure (particularly disorganised) attachment is associated with
various mental health problems for example, conduct disorder (van Ijzendoorn et al.
1999), aggressive and hyperactive behaviour problems (Lyons-Ruth et al. 2009),
depression and anxiety (Lee and Hankin 2009) and childhood posttraumatic stress
symptoms (MacDonald et al. 2008). Approximately 35% of the population will show
insecure attachment (Prior and Glaser 2006).
Classifying attachment
Despite various methods of classifying attachment in childhood, a systematic review
(Lim et al. 2010), concluded that there is a need to develop a new attachment measure
because there is no current tool suitable for use in non-specialist settings (eg in primary
care or paediatric clinics) particularly, as in primary care consultations, where there are
significant time constraints (O'Connor and Byrne 2007).
4
Immunisation and attachment
Attachment behaviours are activated when a child is under stress. Most attachment
assessments place a child into a stressful situation and then evaluate how they use their
primary caregiver to gain comfort. Wilson et al (Wilson et al. 2008) examined the ways
in which health visitors (public health nurses) routinely assess parent/child relationships
and the authors proposed immunisation clinics as a setting where attachment behaviours
could be observed and studied.
A systematic review of psychological interventions for reducing pain and distress during
childhood immunisations sought to determine the efficacy of various psychological
strategies for reducing pain and distress during the procedure (Chambers et al. 2009). It
is clear that children show distress during immunisation, and this distress can be
quantified. Consequently, attachment behaviours may be activated and open to
observation when a child is being immunised.
In this proof-of-concept study, behaviour during immunisation was assessed using a
general measure of parent-child interaction. This approach could allow identification of
candidate characteristics of the parent-child relationship during immunisation which
might provide information on the attachment status of the child.
5
Observing immunisation behaviours
The Mellow Parenting Observational System has been developed as an observational
tool to quantify aspects of the parent-child relationship (Puckering et al. 1994). It was
predicted that behaviours coded on this during immunisation would relate to attachment
behaviours, as parental sensitivity to children’s cues has been related to secure
attachment (Ainsworth et al. 1974). The Mellow Parenting Observational System
records the parent’s reaction to any incidents of child distress and child requests as well
as positive interactions such as playfulness, praise, tone of voice, and physical affect.
Negative responses to distress or other negative interactions are also recorded.
Pain behaviours during immunisation
Children with insecure attachment appear to experience or express higher levels of pain
than children with secure attachments (Porter et al. 2007). Furthermore, research
examining the relationship between parent’s behaviour and how children cope during
painful surgery suggests that children of parents who engage in non-procedural related
talk with their child, who instruct their child to use coping strategies and who direct
humour towards their child, experience less pain during their procedure than children of
parents who do not engage in these behaviours (Blount et al. 1989). Conversely,
children of parents who reassured, apologised to or criticised their child during the
procedure experienced more pain than children of parents who did not engage in these
behaviours. Chambers et al (Chambers et al. 2002) showed that children whose parents
6
use pain promoting behaviours experience greater levels of pain, while pain reducing
behaviours are associated with lower levels of pain.
Current study
To examine whether behaviours during immunisation, coded by both the Mellow
Parenting Observational System and specific pain-related behaviours, relate to
attachment as classified in the Manchester Child Attachment Story Task (MCAST).
These two measures tap into key characteristics of the parent child relationship which
are associated with attachment; warmth and sensitivity as well as a parent’s ability to
modulate their child’s distress.
Method
Participants
Nineteen parent-child pairs took part in the study. One child was removed from the
analysis, because he had intellectual disability and was unable to engage in the MCAST.
Thus data from 18 parent-child pairs were analysed.
7
Materials
Manchester Child Attachment Story Task
The Manchester Child Attachment Story Task (MCAST) is a doll play story stem
technique which measures attachment patterns in middle childhood (Green et al. 2000).
The MCAST works by giving children the beginnings of four stories using a dolls house,
each containing an attachment related theme: the hurt knee, illness, nightmare and
shopping. For example, a child doll – whose mother doll is in the kitchen – is
represented as having stomach ache while watching television in the living room. The
interviewer amplifies the intensity of the doll’s distress until the child is clearly involved
and mildly distressed by what is happening in the scene. At this point the interviewer
hands over to the child saying, “What happens next?” The way the child plays out the
story thereafter is subjected to structured coding based on both Strange Situation
Procedure and Adult Attachment Interview codes and the child is assigned an
attachment classification (Green, Stanley, Smith, & Goldwyn 2000). Children who are
classified as avoidant, ambivalent and disorganised can be grouped together and termed
insecure, resulting in every child being classified as either secure or insecure. In
addition, secure, avoidant and ambivalent can be grouped together and classified as
organised, resulting in every child being classified as either organised or disorganised.
The MCAST has good inter-rater reliability, secure vs. insecure classification (i.e. B vs.
A/C/CC), 94% (Kappa 0.88); categorical D vs. non D classification, 82% (Kappa 0.41).
(Green, Stanley, Smith, & Goldwyn 2000), and shows concurrent validity against other
8
well validated measures of attachment, for example, the Adult Attachment Interview
(AAI) (Goldwyn et al. 2000).
Mellow Parenting Observational System
The Mellow Parenting Observational System (MPOS) is an observational tool to
examine characteristics of the parent-child relationship (Puckering, Rogers, Mills, Cox,
& Mattsson-Graff 1994). This system is an event-sampled observational system that can
be used to describe the interaction between a parent or carer and a child (Wilson et al.
2010). The system covers six domains; anticipation, autonomy, co-operation, warmth
and stimulation and containment of distress each of which has a negative and positive
pole which are statistically independent. These codes meet interrater reliability criteria
greater than 85% (agreement/agreement and disagreement) (Albertsson-Karlgren et al.
2001). Observers trained to use the system must reach inter-rater reliability criteria
greater than 80% (agreement/agreement and disagreement) (Puckering, Rogers, Mills,
Cox, & Mattsson-Graff 1994). The current study used the total positive observational
events with higher scores being associated with more positive behaviours.
9
Pain behaviours
Each immunisation video was assessed for the presence and/or absence of both pain
reducing and pain promoting behaviours as described in Chambers et al (Chambers,
Craig, & Bennett 2002). Pain reducing interactions were described as techniques
adopted by the parent to distract the child through nonprocedural talk, humour directed
to the child, and commands to engage in coping strategies. In comparison, pain
promoting behaviours were described as techniques adopted by the parent which were
designed to be reassuring; providing empathy, apologies or mild criticism, which in turn
gave control to the child. Each video was coded as either having pain reducing or pain
promoting interaction or not having either.
Design and procedure
Participant information sheets and consent forms were sent out to families whose
children were due to receive a pre-school immunisation. Invitations were issued over a
period of 3 months to all eligible families in the participating general medical practice.
These letters were sent out approximately two weeks prior to the appointment.
On the day of their appointment, the researcher approached the parent in the waiting
room and checked that they had received the information sheet, answered any questions
that the parent may have and confirmed whether they gave consent to take part.
10
Following consent the researcher joined the family in the health visitor’s room and
recorded the immunisation procedure with a camcorder. The same personnel, the health
visitor and staff nurse, administered the immunisations to every participant, and they
followed a similar routine with each child. This routine involved getting the child to sit
in the same position; on the parent’s knee, facing and hugging them and giving two
immunisations, one in each arm, at the same time (two children were given the
immunisations separately, one after the other, due to anxiety). The health visitor asked
the child to count to five along with them, telling the child it would be over by the time
they had counted to five. The child was injected when a count of three was reached and
the needle removed when the count had reached five.
The researcher and parent then arranged a suitable time for the researcher to complete
the MCAST with the child. This was never done on the same day as the immunisation
to ensure the child was not still in distress over the immunisation procedure. The
MCAST was conducted either at the GP practice or at the participant’s house; whichever
the parent preferred. Some, but not all, parents wished to remain present during the
MCAST due to the young age of the child.
Following participation, the researcher coded the immunisation tape using the Mellow
Parenting Observation System. As the immunisation videos varied in length, only the
11
minute before and the minute after the immunisation were coded. This was done to
measure how the parent prepared the child for immunisation as well as how they
comforted the child following the procedure. Four (22%) of the immunisation videos
were coded by an additional rater (a psychology student trained to research reliability on
the MPOS), to examine whether behaviours displayed during immunisation could be
coded reliably. These videos were chosen at random using a computer generated
random numbers package. Inter-rater reliability was calculated using Cohen’s Kappa
statistic. In addition the videos were coded for the presence of pain reducing and pain
promoting behaviours.
The MCAST was used to classify attachment and 22% were coded by an additional rater
for inter-rater reliability. Attachment classifications were made following participation
and done on a separate occasion to Mellow Parenting scoring to try and maintain rating
blindness. Attachment groups were compared on immunisation behaviours using tests of
difference between the Mellow Parenting scores and attachment groups as well as Chi-
Square analysis between the attachment groups and the presence and absence of pain
behaviours.
12
Results
Participants
The mean age of the children was 4.12 years (SD 0.44), and the mean age of the parents
was 38.91 years (SD 6.13). Twelve of the children were male. Fourteen mothers and
five fathers accompanied their children to be immunised. The socioeconomic status of
the families varied considerably in the sample, with areas of residence ranging from the
least to most deprived Scottish deprivation categories.
Attachment
The 18 children were given an attachment classification of either secure (50%), avoidant
(27.8%) or disorganised (22.2%). There is also an ambivalent category; however none
of the participants in this sample were classified as ambivalent. Grouping these
classifications together resulted in 50% of the sample being secure, and 50% insecure,
while 77.8% of the sample was organised and 22.2% were disorganised.
As this current study involved using the MCAST on children under the validated age
range of 4-8, the difference between those under age 4 and those over age 4 was
examined. The index of disorganisation was most likely to be affected as
disorganisation can be mimicked by developmental immaturity; however there was no
difference between disorganisation scores of those under and over the age of 4
13
t(15)=0.157, p=0.877, therefore the MCAST seemed an appropriate measure of
attachment in this sample.
Attachment and Mellow Parenting behaviours
The inter-rater reliability of the Mellow Parenting Observational System as measured
using Cohen’s Kappa statistic was 0.82 (excellent inter-rater reliability) (Martin and
Bateson 1993). The current study used the total positive observed events with higher
scores being associated with more positive behaviours. Negative behaviours were too
rare to show any discrimination between subjects. The results showed that the mean
Mellow Parenting score of secure children was 24 (SD =5.77) compared to 22.44
(SD=9.14) for insecure children. The difference was non significant; (t=0.43, df=16,
p=0.67, Cohen’s d= 0.21). The mean Mellow Parenting score for organised children was
24 (SD=7.72), in comparison to 20.5 (SD=6.66) for the disorganised children. This
difference was again non significant; (t=0.82, df=16, p=0.42, Cohen’s d=0.46). Our
pilot data suggests that 170 children would be needed in a definitive study with 80%
power to detect differences between Mellow Parenting scores for children with
organised and disorganised attachment at the 5% significance level.
14
Attachment and pain behaviours
As pain-promoting and pain-reducing interactions have been shown to have an effect on
the level of pain which a child experiences during immunisation (Chambers, Craig, &
Bennett 2002), it was further examined whether these behaviours adopted by the parent
related to attachment classification. Pain promoting behaviours comprised any
behaviours which were designed to be reassuring. These included the parent apologising
to the child or justifying why they needed their immunisation. In this sample, examples
of pain promoting behaviours were parents who focussed on reassuring their children by
repeatedly telling them that it would be “OK”, or that it was not going to be “too sore”.
More specific pain promoting behaviours were direct apology from the parent for taking
their child to be immunised, or explaining the purpose of the immunisation, for example;
“you need this so you’re well to go to school”. These behaviours were considered to
give control to the child but were perhaps not developmentally appropriate for the age
group. Pain reducing behaviours included those which were designed to be distracting,
these included providing the child with a coping strategy or engaging them in
nonprocedural talk. Examples of coping strategies which were observed in this sample
included instructing the child to look out the window while being immunised or to hug
tightly to their parent. Of the parents who engaged in nonprocedural talk, they generally
talked about something the child liked, for example, their favourite television program
or what they were going to be doing after the procedure. Following observation by the
researcher, each video was coded as either containing these behaviours or not containing
them. This observational assessment was based on the behaviour descriptions outlined in
15
Blount et al (Blount, Corbin, Sturges, Wolfe, Prater, & James 1989). The children with
differing attachment classifications were then compared according to the presence or
absence of these behaviours (Table 1).
Table 1 shows that parents of securely attached children engaged in pain reducing
behaviours more than parents of avoidant or disorganised children. In addition, parents
of securely attached children engaged in pain promoting behaviours less often than
parents of avoidant or disorganised children. All parents of securely attached children
engaged in pain reducing behaviours whereas only 33% of parents of insecurely attached
children engaged in these behaviours.
Chi-Square analysis was conducted to test for an association between attachment
classification (secure vs insecure and organised vs disorganised) and parent behaviours
during immunisation (presence or absence of pain reducing and pain-promoting
behaviours). Two cells had expected count less than 5, so an exact significance test was
selected for Pearson’s chi-square. There was a relationship between attachment security
and pain reducing behaviours, (X2(1, N=18) = 9.00, exact p= 0.009). The association
was of moderate strength, φ= .707 and thus attachment security accounted for 49.9% of
the variance in the presence or absence of pain reducing behaviours.
16
Although there was an association between attachment security and pain reducing
behaviours, there was no significant relationship between attachment security and pain
promoting behaviours (X2(1, N=18) = 0.22, exact p= 1.00). There was also no
significant relationship between attachment organisation and pain reducing behaviours
(X2(1, N=18) = 4.02, exact p= 0.08), or attachment organisation and pain promoting
behaviours (X2(1, N=18) = 0.00, exact p= 1.00).
Discussion
In this study, for the first time immunisation was used as an attachment eliciting
paradigm. It was assessed whether immunisation behaviours coded using a general
measure of the quality of parent-child interaction, the Mellow Parenting Observational
System or specific pain promoting and pain reducing behaviours could predict
attachment status as classified in the Manchester Child Attachment Story Task
(MCAST).
Summary of findings
Parents of secure and organised children scored higher in their positive Mellow
Parenting behaviours than parents of insecure and disorganised children, but these
differences were not statistically significant. Parents of securely attached children
engaged in pain reducing behaviours significantly more often than parents of insecure
17
children. They also engaged in pain promoting behaviours less often than parents of
insecure children but this difference was not significant –possibly attributable to a type
II error.
The Mellow Parenting Observational System was expected to relate to attachment as it
taps into key characteristics of the parent child relationship which are associated with
secure attachment, for example, warmth and sensitivity of the parent. The pain
behaviours were measured as they were expected to tap into a parent’s ability to assuage
their child’s distress, which is also associated with attachment. Although these
measurements were examining different aspects of the parent child relationship, there
was some overlap. Reassurance, for example, would be both a pain promoting
behaviour and a positive behaviour depending on the measurement being used. It may be
that during this stressful situation, the parent’s ability to assuage distress is the most
important aspect and this is why pain reducing behaviours showed stronger associations
to the child’s attachment.
Limitations
There are a number of limitations which need to be acknowledged in this study. Piiri et
al (Piira et al. 2007) distinguished between distal and proximal factors involved when a
child undergoes a medical procedure. Distal factors include factors which are present
before the infant attends their immunisation appointment, for example, gender,
18
gestational age, temperament and any early painful experiences, whereas proximal
factors include those which occur immediately before the immunisation, for example,
parental and nurse behaviours. None of the distal factors were investigated in this study
and the health visitor was highly skilled at alleviating the anxiety, and did this uniformly
with each child, thus minimising proximal factors. In addition parental behaviours
could have been affected by the presence of the video camera necessary for completion
of the study. These may have had an effect on how the child reacted while being
immunised and may have minimised differences between the attachment-related
behaviours of parent-child dyads, with the health visitor playing a predominant role.
Attachment behaviours are evident when a child becomes distressed. Immunisation was
considered an experience which would distress all children, however when different
children were observed, it could be seen that children responded with varying levels of
distress. Some children seemed relatively content during the whole procedure, whereas
others were extremely distressed. If the procedure did not put the child into a state of
distress then it is unlikely that the attachment system would have been activated during
the procedure. There are however individual differences between how children
experience or display distress in all attachment measures.
A further limitation of the study is that the analysis was done on a small sample; and
there were few children with disorganised attachment patterns. This may have led to
19
type II errors, disguising genuine associations between attachment and behaviour scores
during immunisation.
Finally, the MCAST is validated for use with children aged 4-8 years. It was the aim to
only use the MCAST on children aged 4 and over, however due to delays getting Health
Service ethical approval, some MCASTs had to be conducted before the child’s fourth
birthday (N=8). In addition, in order to minimise inconvenience to parents, the
researcher offered to conduct the MCASTs either at the family’s general practice or at
their own houses. As parents had varied preferences, testing location varied between
participants. Furthermore, as the children were of such a young age, some parents
requested that they observe the children while the MCAST was being administered.
Although this was discouraged, some parents remained present during its administration
(N=7). It is difficult to consider whether these variations between participants would
have had an effect on the attachment classification of the children.
Implications and future research
A definitive study using the current design would require approximately 170 parent-
child dyads to achieve sufficient statistical power to study Mellow Parenting
observations in a similar design to that reported here. A future study should aim to
video-record at least this number of children while they are being immunised and code
these videos for general parenting, pain reducing and pain promoting behaviours.
20
Developing an attachment measure based on observation of the immunisation procedure
alone has great clinical potential. Attachment could be classified without placing a child
into an artificial state of distress, potentially by nurses or primary care physicians, as
well as providing a measurement which could be used with children of different ages.
The predictive validity for attachment of parent-infant interactions during early infancy
immunisations is certainly worthy of further study. This could offer further potential for
earlier diagnosis and treatment of attachment difficulties.
Conclusions
In conclusion, it was found that parents of children with secure attachments show more
positive behaviours during immunisation as measured by the Mellow Parenting
Observational System but the difference was not significant with our small sample.
Parents of securely attached children engaged in pain reducing behaviours significantly
more often than parents of insecure children. They also engaged in pain promoting
behaviours less frequently than parents of insecure children; however this difference was
not significant. Future research should go on to develop robust, composite measures for
attachment informative behaviours in the immunisation situation and test these in a fully
powered study.
21
References
Ainsworth, M.D.S. 1967. Infancy in Uganda: Infant care and the growth of love Baltimore, John Hopkins University Press.
Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. 1974, "Infant-mother attachment and social development: Socialization as a product of reciprocal responsiveness to signals," In The Integration of a Child into a Social World, M. P. M. Richards, ed., London, UK: Cambridge University Press, pp. 99-135.
Albertsson-Karlgren, U., Graff, M., & Nettelbladt, P. 2001. Mental disease postpartum and parent-infant interaction-evaluation of videotaped sessions. Child Abuse Review, 10, 5-17
Blount, R.L., Corbin, S.M., Sturges, J.W., Wolfe, V.W., Prater, J.M., & James, L.D. 1989. The relationship between adults' behaviour and child coping and distress during BMA/LP procedures: A sequential analysis. Behaviour Therapy, 20, 585-601
Bowlby, J. 1960. Separation anxiety. International Journal of Psychoanalysis, 41, 69-113
Chambers, C.T., Craig, K.D., & Bennett, S.M. 2002. The impact of maternal behaviour on children's pain experiences: An experimental analysis. Journal of Pediatric Psychology, 27, (3) 293-301
Chambers, C.T., Taddio, A., Uman, L.S., & McMurty, C.M. 2009. Psychological interventions for reducing pain and distress during routine childhood immunizations: A systematic review. Clinical Therapeutics, 31, s77-s103
Goldwyn, R., Stanley, C., Smith, V., & Green, J. 2000. The Manchester Child Attachment Story Task: relationship with parental AAI, SAT and child behaviour. Attachment and Human Development, 2, (1) 71-84
Green, J., Stanley, C., Smith, V., & Goldwyn, R. 2000. A new method of evaluating attachment representations in young school age children: The Manchester Child Attachment Story Task. Attachment and Human Development, 2, (1) 48-70
Lee, A. & Hankin, B.L. 2009. Insecure attachment, dysfunctional attitudes, and low self-esteem predicting prospective symptoms of depression and anxiety during adolescence. Journal of Clinical Child and Adolescent Psychology, 38, (2) 219-231
Lim, K., Corlett, L., Law, J., Wilson, P., Gillberg, C., & Minnis, H. 2010. Measuring attachment in large populations: A systematic review. Educational and Child Psychology, 27, (3) 22-32
22
Lyons-Ruth, K., Bureau, J.-F., Riley, C.D., & Atlas-Corbett, A.F. 2009. Socially indiscriminate attachment behavior in the Strange Situation: convergent and discriminate validity in relation to caregiving risk, later behavior problems, and attachment insecurity. Development and Psychopathology, 21, (2) 355-372
MacDonald, H.Z., Beeghly, M., Grant-Knight, W., Augustyn, M., Woods, R.W., Cabral, H., Rose-Jacobs, R., Saxe, G.N., & Frank, D.A. 2008. Longitudinal association between infant disorganized attachment and childhood posttraumatic stress symptoms. Development and Psychopathology, 20, (2) 493-508
Martin, P. & Bateson, P. 1993. Measursing Behaviour-An Introductory Guide, 2nd edition ed. London, UK, Cambridge University Press.
Morgan, G. A. & Ricciuti, H. N. 1969, "Infants' responses to strangers during the first year," In Determinants of Infant Behaviour, IV ed. B. M. Foss, ed., London: Methuen, pp. 253-272.
O'Connor, T.G. & Byrne, J.G. 2007. Attachment measures for research and practice. Child and Adolescent Mental Health, 12, (4) 187-192
Piira, T., Champion, G.D., Bustos, T., Donnelly, N., & Lui, K. 2007. Factors associated with infant pain response following an immunization injection. Human Development, 83, 319-326
Porter, L.S., Davis, D., & Keefe, F.J. 2007. Attachment and pain: Recent findings and futurre directions. Pain, 128, 195-198
Prior, V. & Glaser, D. 2006. Understanding attachment and attachment disorders: Theory, evidence and practice Great Britain, Athenaeum Press.
Puckering, C., Rogers, J., Mills, M., Cox, A.D., & Mattsson-Graff, M. 1994. Progress and evaluation of a group intervention for mothers with parenting difficulties. Child Abuse Review, 3, 299-310
Schaffer, H.R. 1966. The onset of fear of strangers and the incongruity hypothesis. The Journal of Child Psychology and Psychiatry, 7, 95-106
Sroufe, L.A. 2005. Attachment and development: A prospective, longitudinal study from birth to adulthood. Attachment and Human Development, 7, (4) 349-367
van Ijzendoorn, M.H., Schuengel, C., & Bakermans-Kranenburgh, M.J. 1999. Disorganized attachment in early childhood: Meta-analysis of precursors, concomitants, and sequelae. Development and Psychopathology, 11, 225-249
23
Wilson, P., Barbour, R., Graham, C., Currie, M., Puckering, C., & Minnis, H. 2008. Health visitors' assessments of parent-child relationships: a focus group study. International Journal of Nursing Studies, 45, 1137-1147
Wilson, P., Thompson, L., Puckering, C., McConnachie, A., Holden, C., Cassidy, C., & Gillberg C. 2010. Parent-child relationships: are health visitors' judgements reliable? Community Practitioner, 83, (5) 22-25